MONGOCRYPT-755 Implement StrEncode

CMakeLists.txt

@@ -120,6 +120,7 @@ set (MONGOCRYPT_SOURCES
    src/mc-range-encoding.c
    src/mc-rangeopts.c
    src/mc-reader.c
+   src/mc-text-search-str-encode.c
    src/mc-tokens.c
    src/mc-writer.c
    src/mongocrypt-binary.c
@@ -474,6 +475,7 @@ set (TEST_MONGOCRYPT_SOURCES
    test/test-mc-range-mincover.c
    test/test-mc-rangeopts.c
    test/test-mc-reader.c
+   test/test-mc-text-search-str-encode.c
    test/test-mc-tokens.c
    test/test-mc-range-encoding.c
    test/test-mc-writer.c

src/mc-fle2-encryption-placeholder-private.h

@@ -119,6 +119,58 @@ bool mc_FLE2RangeInsertSpec_parse(mc_FLE2RangeInsertSpec_t *out,
                                   bool use_range_v2,
                                   mongocrypt_status_t *status);
 
+typedef struct {
+    // mlen is the max string length that can be indexed.
+    uint32_t mlen;
+    // lb is the lower bound on the length of substrings to be indexed.
+    uint32_t lb;
+    // ub is the upper bound on the length of substrings to be indexed.
+    uint32_t ub;
+} mc_FLE2SubstringInsertSpec_t;
+
+typedef struct {
+    // lb is the lower bound on the length of suffixes to be indexed.
+    uint32_t lb;
+    // ub is the upper bound on the length of suffixes to be indexed.
+    uint32_t ub;
+} mc_FLE2SuffixInsertSpec_t;
+
+typedef struct {
+    // lb is the lower bound on the length of prefixes to be indexed.
+    uint32_t lb;
+    // ub is the upper bound on the length of prefixes to be indexed.
+    uint32_t ub;
+} mc_FLE2PrefixInsertSpec_t;
+
+typedef struct {
+    // v is the value to encrypt.
+    const char *v;
+    uint32_t len;
+
+    // substr is the spec for substring indexing.
+    struct {
+        mc_FLE2SubstringInsertSpec_t value;
+        bool set;
+    } substr;
+
+    // suffix is the spec for suffix indexing.
+    struct {
+        mc_FLE2SuffixInsertSpec_t value;
+        bool set;
+    } suffix;
+
+    // prefix is the spec for prefix indexing.
+    struct {
+        mc_FLE2PrefixInsertSpec_t value;
+        bool set;
+    } prefix;
+
+    // casef indicates if case folding is enabled.
+    bool casef;
+    // diacf indicates if diacritic folding is enabled.
+    bool diacf;
+} mc_FLE2TextSearchInsertSpec_t;
+
 /** FLE2EncryptionPlaceholder implements Encryption BinData (subtype 6)
  * sub-subtype 0, the intent-to-encrypt mapping. Contains a value to encrypt and
  * a description of how it should be encrypted.

src/mc-text-search-str-encode-private.h

@@ -0,0 +1,76 @@
+/*
+ * Copyright 2024-present MongoDB, Inc.
+ *
+ * Licensed under the Apache License, Version 2.0 (the "License");
+ * you may not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ *   http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+#ifndef MONGOCRYPT_TEXT_SEARCH_STR_ENCODE_PRIVATE_H
+#define MONGOCRYPT_TEXT_SEARCH_STR_ENCODE_PRIVATE_H
+
+#include "mc-fle2-encryption-placeholder-private.h"
+#include "mongocrypt-status-private.h"
+#include "mongocrypt.h"
+
+// Represents a valid unicode string with the bad character 0xFF appended to the end. This is our base string which
+// we build substring trees on. Stores all the valid code points in the string, plus one code point for 0xFF.
+// Exposed for testing.
+typedef struct {
+    char *data;
+    uint32_t len;
+    uint32_t *codepoint_offsets;
+    uint32_t codepoint_len;
+} mc_utf8_string_with_bad_char_t;
+
+// Set of substrings of a shared base string.
+typedef struct _mc_substring_set_t mc_substring_set_t;
+
+// Iterator on substring_set.
+typedef struct {
+    mc_substring_set_t *set;
+    uint32_t cur_idx;
+} mc_substring_set_iter_t;
+
+// Point the iterator to the first substring of the given set.
+void mc_substring_set_iter_init(mc_substring_set_iter_t *it, mc_substring_set_t *set);
+
+// Get the next substring, its length, and its count. Returns false if the set does not have a next element, true
+// otherwise.
+bool mc_substring_set_iter_next(mc_substring_set_iter_t *it, const char **str, uint32_t *len, uint32_t *count);
+
+// Result of a StrEncode. Contains the computed prefix, suffix, and substring trees, or NULL if empty, as well as the
+// exact string.
+typedef struct {
+    // Base string which the substring sets point to.
+    mc_utf8_string_with_bad_char_t *base_string;
+    // Set of encoded suffixes.
+    mc_substring_set_t *suffix_set;
+    // Set of encoded prefixes.
+    mc_substring_set_t *prefix_set;
+    // Set of encoded substrings.
+    mc_substring_set_t *substring_set;
+    // Encoded exact string.
+    char *exact;
+    size_t exact_len;
+} mc_str_encode_sets_t;
+
+// Run StrEncode with the given spec.
+mc_str_encode_sets_t *mc_text_search_str_encode(const mc_FLE2TextSearchInsertSpec_t *spec, mongocrypt_status_t *status);
+
+// TODO MONGOCRYPT-759 This helper only exists to test folded_len != unfolded_len; make the test actually use folding
+mc_str_encode_sets_t *mc_text_search_str_encode_helper(const mc_FLE2TextSearchInsertSpec_t *spec,
+                                                       uint32_t unfolded_len,
+                                                       mongocrypt_status_t *status);
+
+void mc_str_encode_sets_destroy(mc_str_encode_sets_t *sets);
+
+#endif /* MONGOCRYPT_TEXT_SEARCH_STR_ENCODE_PRIVATE_H */